PyPDFForm
PyPDFForm copied to clipboard
chinapandaman
PPF-726: Controlling Form Widget Layer
PyPDFForm=1.4.33
Controlling Form Widget Layer
Is there a way to control the layer an added widget occupies? I want preexisting overlapping graphics to appear on top, but the widgets are on top instead.
PDForm.py
import logging
import os
import sys
from collections import defaultdict
from PyPDFForm import PdfWrapper
# Setup logging
logging.basicConfig(level=logging.DEBUG,
format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
datefmt='%Y-%m-%d %H:%M:%S',
filename='pdform.log',
filemode='w')
logger = logging.getLogger(__name__)
class PDForm:
def __init__(self, input_pdf, form_file, form_field_x_offset=0, form_field_y_offset=8, page_height=792):
self.input_pdf = input_pdf
self.form_file = form_file
self.form_field_x_offset = form_field_x_offset
self.form_field_y_offset = form_field_y_offset
self.page_height = page_height
self.output_pdf = self.generate_output_filename()
self.single_line_max_height = 35 # Example constant, adjust as needed
def generate_output_filename(self):
base_name = os.path.splitext(os.path.basename(self.input_pdf))[0]
return f"{base_name}_form.pdf"
def parse_page_range(self, page_range, total_pages):
try:
if '-' in page_range:
start, end = map(int, page_range.split('-'))
return range(start, min(end, total_pages) + 1)
else:
return [min(int(page_range), total_pages)]
except ValueError as e:
logger.error(f"Invalid page range format: {page_range}. Error: {e}")
return []
def read_form_spec(self):
form_spec = defaultdict(list)
with open(self.form_file, 'r') as f:
for line in f:
try:
parts = line.strip().split(',')
if len(parts) != 7:
logger.error(f"Line does not have 7 parts: {line.strip()}")
continue
page_range, field_type, field_name, x, y, width, height = parts
x, y, width, height = map(float, (x, y, width, height))
page_nums = self.parse_page_range(page_range, 100) # Assuming max 100 pages for simplicity
for page_num in page_nums:
form_spec[page_num].append((field_type, field_name, x, y, width, height))
except ValueError as e:
logger.error(f"Error parsing line: {line}. Error: {e}")
return form_spec
def process_pdf(self):
logger.info(f"Starting to process PDF: {self.input_pdf}")
try:
# Step 1: Read form spec file once
form_spec = self.read_form_spec()
# Step 2: Extract pages from the input PDF
pdf_wrapper = PdfWrapper(self.input_pdf)
pages = pdf_wrapper.pages
total_pages = len(pages)
logger.info(f"Total pages in input PDF: {total_pages}")
# Step 3: Create widgets on the extracted pages
for page_num, page in enumerate(pages, start=1):
if page_num in form_spec:
logger.debug(f"Processing page {page_num}")
for field_type, field_name, x, y, width, height in form_spec[page_num]:
# Calculate multiline based on height
multiline = height > self.single_line_max_height
# Make field_name unique for repeated multi-page data
unique_field_name = f"{field_name}_{page_num}"
# Ensure parameters are within valid ranges
if page_num < 1 or page_num > len(pages):
logger.error(f"Invalid page number: {page_num}")
continue
if x < 0 or y < 0 or width <= 0 or height <= 0:
logger.error(f"Invalid widget dimensions: x={x}, y={y}, width={width}, height={height}")
continue
# Transform x-coordinate
# Override in case of any observed form field misalignment
x += self.form_field_x_offset
# Transform y-coordinate (top left origin vs bottom left origin)
y = self.page_height - y - height
# Override in case of any observed form field misalignment
y += self.form_field_y_offset
# Set text fields to multiline if taller than single line
if height > self.single_line_max_height: multiline = True
logger.debug(f"Creating widget: {unique_field_name}, type: {field_type}, on page {page_num} at ({x}, {y}) with size ({width}, {height}), multiline: {multiline}")
try:
# page.create_widget( # Single page attempted work around (that throws exception)
pdf_wrapper.create_widget(
widget_type=field_type,
name=unique_field_name,
# page_number=0, # For single page attempted work around
page_number=page_num,
x=x,
y=y,
width=width,
height=height,
max_length=0, # optional
font="Helvetica", # optional, default "Helvetica"
font_size=10, # optional, default 12
font_color=(0, 0, 0), # optional, default (0, 0, 0)
bg_color=(250, 250, 254), # optional
border_color=(0, 0, 0), # optional
border_width=0, # optional
alignment=0, # optional, 0=left, 1=center, 2=right
multiline=multiline # optional, calculated based on height
)
logger.debug(f"Added field: {field_type} - {unique_field_name} on page 0 multiline: {multiline}")
except ValueError as e:
logger.error(f"Error creating widget for field {field_name} on page 0. Error: {e}")
except IndexError as e:
logger.error(f"Index error creating widget for field {field_name} on page 0. Error: {e}")
except Exception as e:
logger.error(f"Unexpected error creating widget for field {field_name} on page 0. Error: {e}")
# Step 4: Merge the modified pages back into a single PDF (of attempted single page workaround)
# logger.debug("Merging pages into a single PDF")
# merged_pdf = pages[0]
# for i, page in enumerate(pages[1:], start=2):
# logger.debug(f"Merging page {i}/{len(pages)}")
# merged_pdf += page
# Step 5: Save the final merged PDF
logger.debug(f"Saving final merged PDF to {self.output_pdf}")
with open(self.output_pdf, "wb") as final_output_pdf_file:
# final_output_pdf_file.write(merged_pdf.read())
final_output_pdf_file.write(pdf_wrapper.read())
logger.info(f"PDF processing completed. Output saved to {self.output_pdf}")
except FileNotFoundError as e:
logger.error(f"Form file not found: {self.form_file}. Error: {e}")
except Exception as e:
logger.error(f"An error occurred while processing the PDF. Error: {e}")
# Example usage
if __name__ == "__main__":
if len(sys.argv) != 3:
logger.error("Incorrect number of arguments provided")
print('Usage: python script.py <input_pdf> <input_form_file>')
sys.exit(1)
input_pdf = sys.argv[1]
form_file = sys.argv[2]
logger.info(f"Starting PDForm process with input PDF: {input_pdf} and form file: {form_file}")
pd_form = PDForm(input_pdf, form_file)
pd_form.process_pdf()
### Example form_spec.txt file named "THB - 90-Day Reflection Journal - Current_spec.txt"
for a 96 page document in which the same fields are required on pp 5-94
Having the format:
page[-endPage],fieldType,fieldName,x,y,width,height
5-94,text,today,403,59,136,22
5-94,text,feeling,403,107,136,22
5-94,text,grateful1,85,218,465,22
5-94,text,grateful2,85,250,465,22
5-94,text,grateful3,85,282,465,22
5-94,text,reflections,64,406,485,153
5-94,text,wishes,64,604,485,146
### Example command line
$ python PDForm.py "THB - 90-Day Reflection Journal - Current.pdf" "THB - 90-Day Reflection Journal - Current_spec.txt"
Input / Output PDFs
Input PDF: https://drive.google.com/file/d/1co3VmLRQVCYjxwuzwdwiL7FerlgkFqIg/view?usp=sharing Output PDF: https://drive.google.com/file/d/1yLG3C3_GCc0K9gMWjE9GnJYQ4_8UY1vA/view?usp=sharing
Screenshots
Input:
Output:
Hey, thanks for posting.
Unfortunately there's currently no such layer control functionality supported by the library. In your case I have two suggestions:
- If after you created the widgets you are filling using
PdfWrapper, you shouldn't worry about this issue as it will naturally remove widgets and leave only texts on your PDFs after filling. - If you are filling using
FormWrapper, I just released v1.4.35 and alpha band is now supported forbg_color. So you can set a combination ofbg_color=(250, 250, 254, 0)andborder_width=0to make a widget transparent. After which if you setflatten=Truewhen filling you will result in a same filled PDF as if you are filling usingPdfWrapper.
Let me know if you have more questions.
I just updated to the latest release of PyPDFForm and ran it again. The filled widgets are behaving the same as the unfilled ones and still occlude the background images. I am using PdfWrapper to create the widgets. I tried filling a widget populated PDF both before and after the update, and the behaviour was the same. I'm not currently using the FormWrapper method, but if the result when using it the way you suggest should be behave the same as using the PdfWrapper approach, then it might not be worth trying the conversion. The inclusion of an alpha channel for the background does sound promising though. Would it be possible to include that in the PdfWrapper style as well?
Hey sorry for a rather late response. I have been quite busy recently.
So I'm not sure if I understand your issue now. I downloaded your form created using the library and tried this snippet:
from PyPDFForm import PdfWrapper
pdf = PdfWrapper("THB - 90-Day Reflection Journal - Current_form.pdf")
pdf = pdf.fill(
{
"today_5": "foo",
"feeling_5": "foo",
"grateful1_5": "foo",
"grateful2_5": "foo",
"grateful3_5": "foo",
"reflections_5": "foo",
"wishes_5": "foo"
},
)
with open("output.pdf", "wb+") as f:
f.write(pdf.read())
The output seems normal to me?